def put_custom_record(self, environ, coll):
chunks = []
while True:
buff = environ["wsgi.input"].read()
print("LEN", len(buff))
if not buff:
break
chunks.append(buff)
data = b"".join(chunks)
params = dict(parse_qsl(environ.get("QUERY_STRING")))
rec_type = "resource"
headers = {"Content-Type": environ.get("CONTENT_TYPE", "text/plain")}
target_uri = params.get("url")
if not target_uri:
return WbResponse.json_response({"error": "no url"})
timestamp = params.get("timestamp")
if timestamp:
headers["WARC-Date"] = timestamp_to_iso_date(timestamp)
put_url = self.custom_record_path.format(url=target_uri,
coll=coll,
rec_type=rec_type)
res = requests.put(put_url, headers=headers, data=data)
res = res.json()
return WbResponse.json_response(res)
def get_wacz(self, environ, coll):
# if self.pending_count != 0 or self.pending_size != 0:
# return WbResponse.json_response(
# {"error": "not_ready"}, status="404 Not Found"
# )
params = dict(parse_qsl(environ.get("QUERY_STRING")))
archive_dir = os.path.join("collections", coll, "archive")
all_warcs = [
os.path.join(archive_dir, name) for name in os.listdir(archive_dir)
]
all_warcs.append("-o")
all_warcs.append("/tmp/out/archive.wacz")
url = params.get("url")
if url:
all_warcs.append("--url")
all_warcs.append(url)
try:
wacz_main(all_warcs)
except Exception as e:
print(e)
return WbResponse.json_response({"done": "/tmp/out/archive.wacz"})
def serve_listing(self, environ):
result = {
'fixed': self.warcserver.list_fixed_routes(),
'dynamic': self.warcserver.list_dynamic_routes()
}
return WbResponse.json_response(result)
def exit(self, environ=None):
import uwsgi
import signal
resp = WbResponse.json_response({})
os.kill(uwsgi.masterpid(), signal.SIGTERM)
return resp
def test_wbresponse_json_response():
body = dict(pywb=1, wr=2)
res = WbResponse.json_response(body)
status_headers = res.status_headers
assert status_headers.statusline == '200 OK'
assert ('Content-Type',
'application/json; charset=utf-8') in status_headers.headers
assert json.loads(res.body[0]) == body
def put_custom_record(self, environ, coll="$root"):
""" When recording, PUT a custom WARC record to the specified collection
(Available only when recording)
:param dict environ: The WSGI environment dictionary for the request
:param str coll: The name of the collection the record is to be served from
"""
chunks = []
while True:
buff = environ["wsgi.input"].read()
if not buff:
break
chunks.append(buff)
data = b"".join(chunks)
params = dict(parse_qsl(environ.get("QUERY_STRING")))
rec_type = "resource"
headers = {"Content-Type": environ.get("CONTENT_TYPE", "text/plain")}
target_uri = params.get("url")
if not target_uri:
return WbResponse.json_response({"error": "no url"},
status="400 Bad Request")
timestamp = params.get("timestamp")
if timestamp:
headers["WARC-Date"] = timestamp_to_iso_date(timestamp)
put_url = self.put_custom_record_path.format(url=target_uri,
coll=coll,
rec_type=rec_type)
res = requests.put(put_url, headers=headers, data=data)
res = res.json()
return WbResponse.json_response(res)
def serve_listing(self, environ):
"""Serves the response for WARCServer fixed and dynamic listing (paths)
:param dict environ: The WSGI environment dictionary for the request
:return: WbResponse containing the frontend apps WARCServer URL paths
:rtype: WbResponse
"""
result = {'fixed': self.warcserver.list_fixed_routes(),
'dynamic': self.warcserver.list_dynamic_routes()
}
return WbResponse.json_response(result)
def serve_listing(self, environ):
"""Serves the response for WARCServer fixed and dynamic listing (paths)
:param dict environ: The WSGI environment dictionary for the request
:return: WbResponse containing the frontend apps WARCServer URL paths
:rtype: WbResponse
"""
result = {
'fixed': self.warcserver.list_fixed_routes(),
'dynamic': self.warcserver.list_dynamic_routes()
}
return WbResponse.json_response(result)
def put_screenshot(self, environ, coll):
self.ensure_coll_exists(coll)
headers = {'Content-Type': environ.get('CONTENT_TYPE', 'text/plain')}
query_data = parse_qs(environ.get('QUERY_STRING'))
url = query_data.get('target_uri', [])
if url:
url = url[0]
if not url:
return WbResponse.json_response({'error': 'no target_uri'})
put_url = self.screenshot_recorder_path.format(url=url, coll=coll)
res = requests.put(put_url,
headers=headers,
data=environ['wsgi.input'])
res = res.json()
return WbResponse.json_response(res)
def put_record(self, environ, coll, target_uri_format, rec_type, params, data):
self.ensure_coll_exists(coll)
headers = {'Content-Type': environ.get('CONTENT_TYPE', 'text/plain')}
url = params.get('url')
if not url:
return WbResponse.json_response({'error': 'no url'})
timestamp = params.get('timestamp')
if timestamp:
headers['WARC-Date'] = timestamp_to_iso_date(timestamp)
target_uri = target_uri_format.format(url=url)
put_url = self.custom_record_path.format(
url=target_uri, coll=coll, rec_type=rec_type
)
res = requests.put(put_url, headers=headers, data=data)
res = res.json()
return WbResponse.json_response(res)
def test_wbresponse_callable():
expected_body = dict(pywb=1, wr=2)
res = WbResponse.json_response(expected_body)
env = dict(REQUEST_METHOD='GET')
expected_passed_values = dict(status_line='200 OK',
headers=[('Content-Type',
'application/json; charset=utf-8'),
('Content-Length', '17')])
passed_values = dict(status_line=None, headers=None)
def start_response(status_line, headers):
passed_values['status_line'] = status_line
passed_values['headers'] = headers
body = res(env, start_response)
assert json.loads(body[0]) == expected_body
assert passed_values == expected_passed_values
def get_pending(self, environ):
return WbResponse.json_response({
"count": self.pending_count,
"size": self.pending_size
})
def render_content(self, wb_url, kwargs, environ):
wb_url = wb_url.replace('#', '%23')
wb_url = WbUrl(wb_url)
history_page = environ.pop('HTTP_X_WOMBAT_HISTORY_PAGE', '')
if history_page:
wb_url.url = history_page
is_ajax = True
else:
is_ajax = self.is_ajax(environ)
is_timegate = self._check_accept_dt(wb_url, environ)
self.prepare_env(environ)
host_prefix = environ['pywb.host_prefix']
rel_prefix = self.get_rel_prefix(environ)
full_prefix = host_prefix + rel_prefix
pywb_static_prefix = environ['pywb.static_prefix'] + '/'
is_proxy = ('wsgiprox.proxy_host' in environ)
# if OPTIONS in proxy mode, just generate the proxy responss
if is_proxy and self.is_preflight(environ):
return WbResponse.options_response(environ)
if self.use_js_obj_proxy:
content_rw = self.js_proxy_rw
else:
content_rw = self.default_rw
# no redirects if in proxy
redirect_to_exact = self.redirect_to_exact and not is_proxy
# Check Prefer
pref_mod, pref_applied = self._get_prefer_mod(wb_url, environ,
content_rw, is_proxy)
response = None
keep_frame_response = False
# prefer overrides custom response?
if pref_mod is not None:
# fast-redirect to preferred
if redirect_to_exact and not is_timegate and pref_mod != wb_url.mod:
new_url = full_prefix + wb_url.to_str(mod=pref_mod)
headers = [('Preference-Applied', pref_applied),
('Vary', 'Prefer')]
return WbResponse.redir_response(new_url,
'307 Temporary Redirect',
headers=headers)
else:
wb_url.mod = pref_mod
else:
if kwargs.get('output'):
response = self.handle_timemap(wb_url, kwargs, full_prefix)
elif wb_url.is_query():
response = self.handle_query(environ, wb_url, kwargs,
full_prefix)
else:
response = self.handle_custom_response(environ, wb_url,
full_prefix,
host_prefix, kwargs)
keep_frame_response = (not kwargs.get('no_timegate_check')
and is_timegate
and not is_proxy) or redirect_to_exact
if response and not keep_frame_response:
return self.format_response(response, wb_url, full_prefix,
is_timegate, is_proxy)
if is_proxy:
environ['pywb_proxy_magic'] = environ['wsgiprox.proxy_host']
urlrewriter = IdentityUrlRewriter(wb_url, '')
framed_replay = False
else:
urlrewriter = UrlRewriter(wb_url,
prefix=full_prefix,
full_prefix=full_prefix,
rel_prefix=rel_prefix,
pywb_static_prefix=pywb_static_prefix)
framed_replay = self.framed_replay
url_parts = urlsplit(wb_url.url)
if not url_parts.path:
return self.send_redirect('/', url_parts, urlrewriter)
self.unrewrite_referrer(environ, full_prefix)
urlkey = canonicalize(wb_url.url)
inputreq = RewriteInputRequest(environ, urlkey, wb_url.url, content_rw)
inputreq.include_method_query(wb_url.url)
range_start, range_end, skip_record = self._check_range(
inputreq, wb_url)
setcookie_headers = None
cookie_key = None
if self.cookie_tracker:
cookie_key = self.get_cookie_key(kwargs)
if cookie_key:
res = self.cookie_tracker.get_cookie_headers(
wb_url.url, urlrewriter, cookie_key,
environ.get('HTTP_COOKIE', ''))
inputreq.extra_cookie, setcookie_headers = res
r = self._do_req(inputreq, wb_url, kwargs, skip_record)
if r.status_code >= 400:
error = None
try:
error = r.raw.read()
except Exception:
pass
finally:
no_except_close(r.raw)
if error:
error = error.decode('utf-8')
else:
error = ''
details = dict(args=kwargs, error=error)
if r.status_code == 404:
raise NotFoundException(url=wb_url.url, msg=details)
else:
raise UpstreamException(r.status_code,
url=wb_url.url,
details=details)
cdx = CDXObject(r.headers.get('Warcserver-Cdx').encode('utf-8'))
cdx_url_parts = urlsplit(cdx['url'])
if cdx_url_parts.path.endswith(
'/') and not url_parts.path.endswith('/'):
# add trailing slash
new_path = url_parts.path + '/'
no_except_close(r.raw)
return self.send_redirect(new_path, url_parts, urlrewriter)
# only redirect to exact if not live, otherwise set to false
redirect_to_exact = redirect_to_exact and not cdx.get('is_live')
# return top-frame timegate response, with timestamp from cdx
if response and keep_frame_response and (not redirect_to_exact
or not is_timegate):
no_except_close(r.raw)
return self.format_response(response, wb_url, full_prefix,
is_timegate, is_proxy,
cdx['timestamp'])
stream = BufferedReader(r.raw, block_size=BUFF_SIZE)
record = self.loader.parse_record_stream(stream,
ensure_http_headers=True)
memento_dt = r.headers.get('Memento-Datetime')
target_uri = r.headers.get('WARC-Target-URI')
# cdx['urlkey'] = urlkey
# cdx['timestamp'] = http_date_to_timestamp(memento_dt)
# cdx['url'] = target_uri
set_content_loc = False
# Check if Fuzzy Match
if target_uri != wb_url.url and cdx.get('is_fuzzy') == '1':
set_content_loc = True
# if redirect to exact timestamp (only set if not live)
if redirect_to_exact:
if set_content_loc or is_timegate or wb_url.timestamp != cdx.get(
'timestamp'):
new_url = urlrewriter.get_new_url(url=target_uri,
timestamp=cdx['timestamp'],
mod=wb_url.mod)
resp = WbResponse.redir_response(new_url,
'307 Temporary Redirect')
if self.enable_memento:
if is_timegate and not is_proxy:
self._add_memento_links(target_uri,
full_prefix,
memento_dt,
cdx['timestamp'],
resp.status_headers,
is_timegate,
is_proxy,
pref_applied=pref_applied,
mod=pref_mod,
is_memento=False)
else:
resp.status_headers['Link'] = MementoUtils.make_link(
target_uri, 'original')
return resp
self._add_custom_params(cdx, r.headers, kwargs, record)
if self._add_range(record, wb_url, range_start, range_end):
wb_url.mod = 'id_'
if is_ajax:
head_insert_func = None
urlrewriter.rewrite_opts['is_ajax'] = True
else:
top_url = self.get_top_url(full_prefix, wb_url, cdx, kwargs)
head_insert_func = (self.head_insert_view.create_insert_func(
wb_url,
full_prefix,
host_prefix,
top_url,
environ,
framed_replay,
coll=kwargs.get('coll', ''),
replay_mod=self.replay_mod,
metadata=kwargs.get('metadata', {}),
config=self.config))
cookie_rewriter = None
if self.cookie_tracker and cookie_key:
# skip add cookie if service worker is not 200
# it seems cookie headers from service workers are not applied, so don't update in cache
if wb_url.mod == 'sw_':
cookie_key = None
cookie_rewriter = self.cookie_tracker.get_rewriter(
urlrewriter, cookie_key)
urlrewriter.rewrite_opts['ua_string'] = environ.get('HTTP_USER_AGENT')
result = content_rw(record, urlrewriter, cookie_rewriter,
head_insert_func, cdx, environ)
status_headers, gen, is_rw = result
if history_page:
title = DefaultRewriter._extract_title(gen)
if not title:
title = unquote(environ.get('HTTP_X_WOMBAT_HISTORY_TITLE', ''))
if not title:
title = history_page
self._add_history_page(cdx, kwargs, title)
return WbResponse.json_response({'title': title})
if setcookie_headers:
status_headers.headers.extend(setcookie_headers)
if ' ' not in status_headers.statusline:
status_headers.statusline += ' None'
if not is_ajax and self.enable_memento:
self._add_memento_links(cdx['url'],
full_prefix,
memento_dt,
cdx['timestamp'],
status_headers,
is_timegate,
is_proxy,
cdx.get('source-coll'),
mod=pref_mod,
pref_applied=pref_applied)
set_content_loc = True
if set_content_loc and not redirect_to_exact and not is_proxy:
status_headers.headers.append(
('Content-Location',
urlrewriter.get_new_url(timestamp=cdx['timestamp'],
url=cdx['url'])))
if not is_proxy:
self.add_csp_header(wb_url, status_headers)
response = WbResponse(status_headers, gen)
if is_proxy and environ.get('HTTP_ORIGIN'):
response.add_access_control_headers(environ)
if r.status_code == 200 and kwargs.get(
'cache') == 'always' and environ.get('HTTP_REFERER'):
response.status_headers[
'Cache-Control'] = 'public, max-age=31536000, immutable'
return response
def render_content(self, wb_url, kwargs, environ):
wb_url = wb_url.replace('#', '%23')
wb_url = WbUrl(wb_url)
proto = environ.get('HTTP_X_FORWARDED_PROTO', self.force_scheme)
if proto:
environ['wsgi.url_scheme'] = proto
history_page = environ.pop('HTTP_X_WOMBAT_HISTORY_PAGE', '')
if history_page:
wb_url.url = history_page
is_ajax = True
else:
is_ajax = self.is_ajax(environ)
is_timegate = self._check_accept_dt(wb_url, environ)
host_prefix = self.get_host_prefix(environ)
rel_prefix = self.get_rel_prefix(environ)
full_prefix = host_prefix + rel_prefix
environ['pywb.host_prefix'] = host_prefix
pywb_static_prefix = host_prefix + environ.get(
'pywb.app_prefix', '') + environ.get('pywb.static_prefix',
'/static/')
is_proxy = ('wsgiprox.proxy_host' in environ)
response = self.handle_custom_response(environ, wb_url, full_prefix,
host_prefix, kwargs)
if response:
return self.format_response(response, wb_url, full_prefix,
is_timegate, is_proxy)
if is_proxy:
environ['pywb_proxy_magic'] = environ['wsgiprox.proxy_host']
urlrewriter = IdentityUrlRewriter(wb_url, '')
framed_replay = False
else:
urlrewriter = UrlRewriter(wb_url,
prefix=full_prefix,
full_prefix=full_prefix,
rel_prefix=rel_prefix,
pywb_static_prefix=pywb_static_prefix)
framed_replay = self.framed_replay
url_parts = urlsplit(wb_url.url)
if not url_parts.path:
return self.send_redirect('/', url_parts, urlrewriter)
self.unrewrite_referrer(environ, full_prefix)
urlkey = canonicalize(wb_url.url)
if self.use_js_obj_proxy:
content_rw = self.js_proxy_rw
else:
content_rw = self.default_rw
inputreq = RewriteInputRequest(environ, urlkey, wb_url.url, content_rw)
inputreq.include_method_query(wb_url.url)
range_start, range_end, skip_record = self._check_range(
inputreq, wb_url)
setcookie_headers = None
cookie_key = None
if self.cookie_tracker:
cookie_key = self.get_cookie_key(kwargs)
if cookie_key:
res = self.cookie_tracker.get_cookie_headers(
wb_url.url, urlrewriter, cookie_key,
environ.get('HTTP_COOKIE', ''))
inputreq.extra_cookie, setcookie_headers = res
r = self._do_req(inputreq, wb_url, kwargs, skip_record)
if r.status_code >= 400:
error = None
try:
error = r.raw.read()
except Exception:
pass
finally:
no_except_close(r.raw)
if error:
error = error.decode('utf-8')
else:
error = ''
details = dict(args=kwargs, error=error)
raise UpstreamException(r.status_code,
url=wb_url.url,
details=details)
cdx = CDXObject(r.headers.get('Warcserver-Cdx').encode('utf-8'))
cdx_url_parts = urlsplit(cdx['url'])
if cdx_url_parts.path.endswith(
'/') and not url_parts.path.endswith('/'):
# add trailing slash
new_path = url_parts.path + '/'
no_except_close(r.raw)
return self.send_redirect(new_path, url_parts, urlrewriter)
stream = BufferedReader(r.raw, block_size=BUFF_SIZE)
record = self.loader.parse_record_stream(stream,
ensure_http_headers=True)
memento_dt = r.headers.get('Memento-Datetime')
target_uri = r.headers.get('WARC-Target-URI')
# cdx['urlkey'] = urlkey
# cdx['timestamp'] = http_date_to_timestamp(memento_dt)
# cdx['url'] = target_uri
set_content_loc = False
# Check if Fuzzy Match
if target_uri != wb_url.url and cdx.get('is_fuzzy') == '1':
set_content_loc = True
# if redir to exact, redir if url or ts are different
if self.redirect_to_exact:
if (set_content_loc or (wb_url.timestamp != cdx.get('timestamp')
and not cdx.get('is_live'))):
new_url = urlrewriter.get_new_url(url=target_uri,
timestamp=cdx['timestamp'],
mod=wb_url.mod)
resp = WbResponse.redir_response(new_url,
'307 Temporary Redirect')
if self.enable_memento:
if is_timegate and not is_proxy:
self._add_memento_links(target_uri, full_prefix,
memento_dt, cdx['timestamp'],
resp.status_headers,
is_timegate, is_proxy)
else:
resp.status_headers['Link'] = MementoUtils.make_link(
target_uri, 'original')
return resp
self._add_custom_params(cdx, r.headers, kwargs, record)
if self._add_range(record, wb_url, range_start, range_end):
wb_url.mod = 'id_'
if is_ajax:
head_insert_func = None
urlrewriter.rewrite_opts['is_ajax'] = True
else:
top_url = self.get_top_url(full_prefix, wb_url, cdx, kwargs)
head_insert_func = (self.head_insert_view.create_insert_func(
wb_url,
full_prefix,
host_prefix,
top_url,
environ,
framed_replay,
coll=kwargs.get('coll', ''),
replay_mod=self.replay_mod,
config=self.config))
cookie_rewriter = None
if self.cookie_tracker and cookie_key:
# skip add cookie if service worker is not 200
# it seems cookie headers from service workers are not applied, so don't update in cache
if wb_url.mod == 'sw_':
cookie_key = None
cookie_rewriter = self.cookie_tracker.get_rewriter(
urlrewriter, cookie_key)
urlrewriter.rewrite_opts['ua_string'] = environ.get('HTTP_USER_AGENT')
result = content_rw(record, urlrewriter, cookie_rewriter,
head_insert_func, cdx, environ)
status_headers, gen, is_rw = result
if history_page:
title = DefaultRewriter._extract_title(gen)
if not title:
title = unquote(environ.get('HTTP_X_WOMBAT_HISTORY_TITLE', ''))
if not title:
title = history_page
self._add_history_page(cdx, kwargs, title)
return WbResponse.json_response({'title': title})
if setcookie_headers:
status_headers.headers.extend(setcookie_headers)
if ' ' not in status_headers.statusline:
status_headers.statusline += ' None'
if not is_ajax and self.enable_memento:
self._add_memento_links(cdx['url'], full_prefix, memento_dt,
cdx['timestamp'], status_headers,
is_timegate, is_proxy,
cdx.get('source-coll'))
set_content_loc = True
if set_content_loc and not self.redirect_to_exact:
status_headers.headers.append(
('Content-Location',
urlrewriter.get_new_url(timestamp=cdx['timestamp'],
url=cdx['url'])))
if not is_proxy:
self.add_csp_header(wb_url, status_headers)
response = WbResponse(status_headers, gen)
return response
def page_search(self, environ, coll):
params = dict(parse_qsl(environ.get('QUERY_STRING')))
result = self.solr_ingester.query_solr(coll, params)
return WbResponse.json_response(result)
